Help on syntax for linear mixed models

[Phil Hill]

Dear all,

I am hoping for a bit of guidance on the analysis I am intending to run for my cross sectional data, I have to say I am very new to STATA so apologies in advance, but it has been proposed that I use linear mixed effects modelling. Firstly, I have struggled to find the difference between the mixed and the xtmixed command. Secondly, I would hugely appreciate if someone is able to confirm that the syntax for the analyses I am intending to run would be correct, the second of which has been suggested to me.

The intention of the first analysis is to run a linear mixed effects model to identify the influence the cross-sectional data of individual sport skills, proficiency in sports skills, gender and age on physical activity, with individual participants as random effects, and sports skills, proficiency in sports skills, gender and age as fixed effects. Would the syntax below be appropriate for this?

mixed PA Skill c.skill_proficiency Sex c.age || ID:

The second analysis is interested in the influence of sex, the interaction of maturity and age, and physical competence on throw velocity. A colleague presented this as an option, but I am interested to know how this differs from the
first analysis (mixed compared to xt mixed) and whether the second analysis would include any random-effects or not.

xtmixed throw_vel i.sex i.maturity##c.age c.phys_c

More than happy to provide any further detail. Many thanks.

[Rich Goldstein]

your situation is not completely clear to me but, (1) -xtmixed- is the old name for what is now called -mixed- (if you have an up-to-date version of Stata, just type "help xtmixed"); (2) your second code does not include any random effects (if I remember the old syntax correctly)

[Tiago Pereira]

Not clear why you need a random-effects model.